Juan Caamaño
Software Engineer at AMD
Lyon, Auvergne-Rhône-Alpes, France
Summary
🤩
Rockstar🎓
Top SchoolJuan Caamaño is a compiler-focused software engineer with 12 years of experience bridging research and production-grade toolchains, currently working on the AMDGPU backend at AMD in Lyon. He brings deep LLVM expertise from both industry and academia—leading a protection-suite team at Quarkslab and contributing substantive backend improvements to the high-profile llvm/llvm-project repository. His PhD work and early internships specialized in dynamic code generation and automatic parallelization, informing pragmatic optimizations like dynamic LDS handling and peephole/branch-weight changes for AMD hardware. Comfortable as both a hands-on engineer and team lead, he combines low-level code-generation know-how with a track record of shipping cross-platform, security-minded compiler features.
12 years of coding experience
7 years of employment as a software developer
Doctor of Philosophy (Ph.D.) Ciencias de la computación, Doctor of Philosophy (Ph.D.) Ciencias de la computación at University of StrasbourgMaster's degree Ciencias de la computación, Master's degree Ciencias de la computación at University of Buenos Aires
Spanish, English, French
Stackoverflow
343reputation
93kreached
15answers
0questions
Github Skills (13)
code-generation10
c-language10
amdgpu10
cprogramming-language10
llvm10
optimization9
control-flow8
flow-control8
clang6
vector6
compiler-optimization6
macos6
llvm-ir6
Programming languages (6)
Github contributions (5)
llvm/llvm-project
Apr 2021 - Jul 2026
The LLVM Project is a collection of modular and reusable compiler and toolchain technologies.
Role in this project:
Back-end Developer
Back-end DeveloperContributions:234 reviews, 161 PRs, 173 pushes in 5 years 3 months
Contributions summary:Juan primarily contributed to the AMDGPU backend within the LLVM project. Their work involved modifying and removing flags related to structurizer, a control flow optimization, and implementing changes to support dynamic LDS (Local Data Share) interaction. Furthermore, they made changes to the pre-emit peephole optimization pass and branch weights metadata. These changes aimed to improve code generation for AMDGPU hardware.
compilerllvmtoolchain
jmmartinez/llvm-project
Sep 2023 - Nov 2025
The LLVM Project is a collection of modular and reusable compiler and toolchain technologies.
Contributions:154 pushes, 83 branches in 2 years 2 months
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.